INDEX

Explanations

keywords related to computer hardware brands and prefixes

oai_token-act-pair · gpt-3.5-turbo

numeric data and specific terms related to measurement or classification

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

TAG

-0.53

 academ

-0.51

quit

-0.50

 sidx

-0.47

 IMAGES

-0.47

akeru

-0.47

MET

-0.46

INGTON

-0.46

leck

-0.45

 seminars

-0.45

POSITIVE LOGITS

oxide

0.74

ibrary

0.66

ioxide

0.58

arine

0.54

lycer

0.53

alien

0.53

aternity

0.51

eanor

0.50

plane

0.50

iky

0.50

Activations Density 1.435%

keywords related to computer hardware brands and prefixes

numeric data and specific terms related to measurement or classification

No Comments

No Known Activations